Franco's Blog

• clearnet: https://blog.franco.net.eu.org/
• onion: http://usggwz6vwth276fdyccskoba5u2jis5ugv2ew2fy2blowgf2dv73axad.onion

Table of contents

• Franco's Blog
  • Table of contents
  • Blog content copyright and license
• Documentation for blog.franco.net.eu.org
  • Issue tracker
  • Philosophy
  • Dependencies
  • Installation
    • Generic
    • Debian GNU/Linux
  • Upgrading
  • Building and serving
    • Locally
    • Self-hosted
    • GitLab Pages
  • Tutorial
    • Media files
      • Overview
      • File locations
      • Possible combinations
      • Calling files from another post
      • Notice and warning
    • Static pages
    • Tags and categories
      • Overview
      • The tags score system
      • Tags page
      • Tags in posts
    • Excerpts
    • Anchor headings
    • RSS feeds
    • Search Engine Optimizaition (SEO)
      • meta and other SEO tags
      • Sitemap
      • robots.txt
    • Compressed HTML output
    • Open Graph Protocol and Microformats
    • Comments
    • HTML validation
  • Useful resources
  • Copyright and License
    • Liquid, YAML, scripts, etc...
    • CSS code
  • Changelog and trusted source
  • Crypto donations

Blog content copyright and license

Copyright (C) Franco Masotti

CC-BY-SA 4.0

Documentation for blog.franco.net.eu.org

Instead of maintaning a separate repository for the theme it is simpler for me to put the documentation here.

Issue tracker

https://software.franco.net.eu.org/frnmst/blog/issues

You can use your GitHub account to interact. Have a look here

Philosophy

• No Javascript is used. See The JavaScript Trap.
• No local fonts are served.
• No remote fonts are included (called).
• GitHub's primer theme CSS is used.

Dependencies

The theme is known to work with the the packages listed here and they must be installed on your system.

Package	Executable	Version command	Package Version
Ruby	/usr/bin/ruby	$ ruby --version	ruby 3.1.2p20 (2022-04-12 revision 4491bb740a) [x86_64-linux-gnu]
Bundler	/usr/bin/bundle	$ bundle --version	Bundler version 2.3.15
GNU Make	/usr/bin/make	$ make --version	GNU Make 4.3
pip	/usr/bin/pip	$ pip --version	pip 23.0.1

Installation

Generic

$ gem update
$ gem install bundle
$ make install-dev

Debian GNU/Linux

# apt-get install ruby-bundler
$ make install-dev

Upgrading

Run $ make install

Building and serving

Locally

Local serving, on 127.0.0.1

$ make

Global serving, on 0.0.0.0

$ make serve-global

Connect to http://127.0.0.1:5555/the-flux-of-thought/.

Self-hosted

See https://blog.franco.net.eu.org/post/new-hosting.html

GitLab Pages

If you use GitLab Pages put the following in ./.gitlab-ci.yml. See also https://gitlab.com/pages/jekyll

image: ruby:2.6

test:
  stage: test
  script:
  - gem install 'jekyll:4.0.0' jekyll-sitemap 'forwardable:1.2.0'
  - jekyll build -d test
  artifacts:
    paths:
    - test
  except:
  - master

pages:
  stage: deploy
  script:
  - gem install 'jekyll:4.0.0' jekyll-sitemap 'forwardable:1.2.0'
  - jekyll build -d public
  artifacts:
    paths:
    - public
  only:
  - master

Tutorial

Media files

Extending these ideas I came up with the following:

Overview

Each media file is contained in one directory corresponding to a post. Media content may be referenced from more than one post.

File locations

If out example post is ./_posts/2017-02-11-another-post.md and we want to add media files to it, we must create a corresponding directory in _media:

./_media/2017-02-11-another-post

We can now place our file inside, for example:

./_media/2017-02-11-another-post/terminal.png

Possible combinations

Now, let's go back to ./_posts/2017-02-11-another-post.md. To be able to display that picture we need to use the include liquid tag. In the simplest form the only required parameter is the file name:

{% include image.html file="terminal.png" %}

You can also use the alt, capiton and width tags:

{% include image.html file="terminal.png" alt="ter" caption="A terminal example caption" width="600px" %}

width defaults to 300px if not specified.

You can also use plain markdown, which is not advisable in this case, since you need to input the whole path:

![ter]({{ site.rooturl }}/media/2017-02-11-another-post/terminal.png)

Calling files from another post

In some cases you may want to recall media files from another post. You can do that using the otherpost tag. Let's say we are in the ./_posts/2015-09-09-download-this-theme.md post.

{% include image.html file="2017-02-11-another-post/terminal.png" alt="Terminal" caption="A terminal image from the other post" otherpost=true %}

As you can see you don't have to specify the full path but only the post name slash the file name. The rest is filled in automatically.

Finally, notice the otherpost=true tag at the end.

Notice and warning

If you do not include the alt tag a default one will be provided instead.

This include feature is currently available only for images (img HTML tag).

Static pages

Reading this gave me the idea to move the static pages like index, 404, etc.. in the ./_pages directory. If you want to add new pages remember to add a sensible permalink for each page and remember to keep include: [ "_pages" ] and:

collections:
  pages:
    output: false

in the _config.yml file.

Tags and categories

Overview

In this theme tags and categories are considered the same thing. I will use the word tag to speak about both tags and categories.

I gathered some ideas from here and using the official Jekyll documentation.

A list of tags is present on the home page, i.e: ./_pages/index.html. This list is just a set of links pointing to the appropriate entry in the ./_pages/tags.md page.

The tags score system

A list of tags is present on the home page, i.e: ./_pages/index.html. This list is just a set of links pointing to an appropriate entry in the ./_pages/tags.md page, which are presented using a score criteria. All occurrencies of a tag are counted globally. If that frequency is greater or equal than the tags.score.min variable in the _config.yml file then the tag list element is showed along with its score. On the other hand if the frequency is less than tags.score.min the tag will not be shown. If you want all the tags to be shown, simply set tags.score.min to 1.

You can limit the number of tags from home and tags page by setting the tags.score.link_limit variable to a positive integer number. If you set it false the tags will not be limited.

You may have noticed that the tag list is presented in a descreasing number of occurrencies per tags. This method was chosen so that the relevant content is shown first.

Sorting is done first by tag score and then alphabetically.

Some elements of logic were taken from here, here, here, here and the official Liquid documentation.

Tags page

The purpose of this page is to present links to posts for each tag in the website. This page reports all tags in the website without the tags score system except that the list will be sorted with descreasing frequency. The tags reported on the top of the page follow the tags score system.

Below the tags list there are links to each posts, organized by tags. If a user clicks on one of the links in the tags list, the viewport will be moved to the list of posts containing that tag.

Tags in posts

If a post contains at least one tag, this will be reported at the top of the page with a link pointing to the appropriate bookmark of ./_pages/tags.md (just like in the home page). A user looking at a post can see all related posts just by clicking at those tag links.

The tag entry is optional which means that you can specify from 0 to n tags. If you want to add some, simply write your comma separated list of tags in the front matter of your post:

tags: [tag 0, tag 1, ..., tag n]

Let's make an example. What follows is the front matter of the ./_posts/2015-09-09-download-this-theme.md post.

---
title: Download this Theme
updated: 2017-04-09 00:29
categories:  Download
tags: [download, terminal, image from another post]
---

As you may have noticed, tags can contain spaces.

Once you run make, you should see the following links just below the Download this Theme title:

*download* *terminal* *image from another post*

Excerpts

An excerpt is a portion of a post reported in the home page.

Any content between <!--excerpt_start--> and <!--excerpt_end--> tags in a post will be used as excerpt.

If one or both of these tags are omitted, the first 20 words of the post will be reported instead. You can specify the number of words in the excerpt by editing the excerpts.words variable in the ./_config.yml file.

If you are not interested in excerpts set the excerpts.enabled variable to false in ./_config.yml.

The liquid code that handles the excerpts was inspired by this article. Here is its license and copyright:

The MIT License (MIT)

Copyright (c) 2015 Joshua Beam

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.

Anchor headings

Anchor headings are clickable links beside a <h[1-6]> tag. The purpose of these elements is to get a quick link to a specific part of a page.

You can modify the appearence of these links from the ./_config.yml file through the anchor_headings variables.

All this is possible thanks to the Jekyll Pure Liquid Heading Anchors library.

RSS feeds

This theme contains an Atom file useful for RSS feeds in ./_pages/feed.xml.

The link to feed.xml is available in the navigation bar of every page. The purpose is that your readers open this link with a feed reader so that they can remain up to date with your posts.

I use these Apache rules in a virtual host directive to redirect common URLs for RSS feeds:

Redirect 301 /feed/ https://blog.franco.net.eu.org/feed.xml
Redirect 301 /rss/ https://blog.franco.net.eu.org/feed.xml
Redirect 301 /feed https://blog.franco.net.eu.org/feed.xml
Redirect 301 /rss https://blog.franco.net.eu.org/feed.xml

Search Engine Optimizaition (SEO)

I'm a noob in this field so I followed this tutorial.

• meta
• sitemaps
• images
  • always use alt tags. For images in ./_assets these are already coded in.
• responsiveness
  • the theme is reported responsive by Google
• robots.txt

meta and other SEO tags

meta tags are available in the ./_includes/head.html file. These include: title, description and canonical links.

The description meta tag uses the first available, in this order, of the following:

• Front matter description tag
• Page excerpt
• Site description

It is advised to use the description tag since excerpts or the site description may not relate to the (whole) page content. Here's an example:

---
title: This is another post
updated: 2017-04-09 23:00
tags: [other, liquid, terminal]
description: How to use markdown and liquid to render images belonging to a
post.
---

You can also edit the meta variable in ./_config.ymlto limit the number of words in the description meta tag. Avoid values greater than 160 as reported by previously cited articles:

Sitemap

A sitemap used for web crawlers is available thanks to the jekyll-sitemap plugin.

There is also a (very) basic sitemap for users under the /sitemap/ permalink which has a link for it at the top of each page.

robots.txt

This very important file must be put under the root of your domain. It is used by web crawlers (a.k.a bots, spiders, etc...) to get access information about the website. In our case it is also important to tell these programs that our site has a sitemap. Read this and this for more information.

For example, since this project is called the-plain-libre and is published under https://frnmst.github.io/the-plain-libre, we need to put the robots.txt file just under https://frnmst.github.io/, i.e

https://frnmst.github.io/robots.txt

Since I didn't have neither a "user page" nor a domain I created a dummy user page repository and I copied the robots.txt file from the generated ./_site directory from Jekyll. This works for me. This means that in my case the robots.txt in this repository is ignored by the bots.

I guess that if you have multiple repository pages you can edit the robots.txt file appropriately and set rules (and sitemap locations) for each one of them.

Compressed HTML output

By default Jekyll leaves lots of whitespace in the html source. You can verify this yourself.

Since useless white space means useless data transmitted, loading the website requires more time.

For this reason I decided to use a compressed HTML output thanks to jekyll-compress-html, which is a pure liquid layout to be used as a wrapper on the ./_layouts/default.html layout.

jekyll-compress-html is released under the MIT license.

Unfortunately this layout has a bug that removes some closing HTML tags such as li. This bug messes up the output, so I decided to look for other solutions and found the whitespace control of the liquid template language. This is not as good as "compressing" the output but it's better than nothing.

Open Graph Protocol and Microformats

See this post for Open Graph Protocol.

I followed this tutorial to implement microformats.

Comments

Done using Isso.

HTML validation

This Jekyll theme has been checked with the HTML5 W3C validator

Useful resources

• https://www.w3.org/
• Especially useful for CSS examples: https://www.w3schools.com/default.asp

Copyright and License

Liquid, YAML, scripts, etc...

The MIT License (MIT)

Copyright (c) 2015 Heiswayi Nrird

Copyright (c) 2017-2023 Franco Masotti

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

CSS code

The MIT License (MIT)

Copyright (c) 2016-2017 GitHub, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Changelog and trusted source

You can check the authenticity of new releases using my public key.

Changelogs, instructions, sources and keys can be found at blog.franco.net.eu.org/software

Crypto donations

• Buy Me a Coffee
• Liberapay
• Bitcoin: bc1qnkflazapw3hjupawj0lm39dh9xt88s7zal5mwu
• Monero: 84KHWDTd9hbPyGwikk33Qp5GW7o7zRwPb8kJ6u93zs4sNMpDSnM5ZTWVnUp2cudRYNT6rNqctnMQ9NbUewbj7MzCBUcrQEY
• Dogecoin: DMB5h2GhHiTNW7EcmDnqkYpKs6Da2wK3zP
• Vertcoin: vtc1qd8n3jvkd2vwrr6cpejkd9wavp4ld6xfu9hkhh0